This dissertation proposes the use of advanced time-varying approaches for modeling the dynamics of the multipath channel in wireless communication networks. These advanced time-varying approaches include linear Kalman innovation models in observable block companion form, and neural network-based models. The e˙ectiveness of these type of models is evaluated through three case studies. The ﬁrst case study involves the identiﬁcation of a linear time-varying Kalman innovation model, for describing measured received signal strength (RSSI) as a function of the speed of the link in an indoor multipath wireless channel. Results for this ﬁrst case study show that the model exhibits both accuracy and robustness. The second case study evaluates the suitability of using a linear time-varying Kalman innovation model of the RSSI, for secret key generation in the physical layer of multipath wireless channels. It was found that the residuals of the Kalman model, due to their signiﬁcant randomness, exhibit a notable potential for secret key generation; indeed, improved values of maximum channel capacity for secret key generation were achieved. At last, the third case study includes the identiﬁcation of a neural network-based autoregressive moving average with exogenous inputs (NN-ARMAX) model and of a neural network-based autoregressive with exogenous inputs (NN-ARX) model, for describing traÿc in a 4G-LTE network. Both models showed similar performance, but the NN-ARMAX has the advantage that it can be converted to a linear time-varying Kalman innovation model, and thus can be used for the implementation of advanced strategies for controlling the operation of the network.